What Does It Mean to Be a
Nonprofit Educational
Measurement Organization
in the 21st Century?
by Randy Elliot Bennett
ForewOrd
In this paper, ETS Distinguished Presidential Appointee Randy Bennett asks what it means to
be a nonprofit educational measurement organization in today’s world. That question is terribly
important because, throughout the social service sector, the line between what nonprofit and forprofit organizations do is becoming blurred. This fuzziness is in part the result of financial pressures
created by decreased government support for social service institutions, forcing those organizations
to become more competitive with one another. That decrease in government support also forces these
organizations to find new ways to generate the funds needed to achieve their missions, driving some
nonprofits to compete with for-profit corporations. The commercial sector, in turn, has realized that
profit can be made from social service, which adds further competitive pressure to the mix.
In this paper, Bennett uses ETS as a case study. He begins by reviewing the federal tax code relating
to educational nonprofit organizations. He then analyzes the circumstances that led to ETS’s founding.
Finally, he offers thoughts on how that tax code and organizational history might be used to guide ETS
in making decisions about its future.
In the end, Bennett presents much more than a historical account. His work is meant to spark a
discussion about how to strike the right balance between upholding and advancing the scientific
principles that undergird the field of educational measurement, on the one hand, and the need to act
in a businesslike manner, on the other.
The paper’s central question — how to remain financially viable in a highly competitive business
environment, while remaining true to the mission of serving the public good — is relevant to all of
those with an interest in the institution of educational measurement.
Regards,
Kurt M. Landgraf
President and CEO
ETS
i
ii
Table of Contents
Abstract..................................................................................................................................................... iv
Acknowledgments.................................................................................................................................... iv
What Does It Mean to Be a Nonprofit Educational Measurement Organization in the 21st Century?..... 1
What Is an Educational Nonprofit?............................................................................................................ 1
Where Did ETS Come From?.................................................................................................................... 3
What Does the Past Imply for the Future?................................................................................................. 8
Summary.................................................................................................................................................. 11
References................................................................................................................................................ 12
iii
ABSTRACT
This paper explores the meaning of nonprofit status in the field of educational measurement.
Educational Testing Service (ETS), the largest nonprofit educational measurement organization,
is used as a case study. The paper reviews the historical roots of ETS from two perspectives. First,
the requirements and history of the Internal Revenue Code governing the establishment and operation
of 501(c)(3) organizations are described. Next, the people and events leading to the establishment
of ETS as a nonprofit educational measurement organization are explored. Finally, the principles
underlying Section 501(c)(3) and those of ETS’s progenitors are brought together to suggest what
it means to be a nonprofit educational measurement organization in the 21st century.
ACKNOWLEDGMENTS
I am grateful to the following people for providing information, comments, or other assistance to
me in developing the presentation on which this paper was based: Isaac Bejar, Michal Beller, Julie
Duminiak, Marisa Farnum, Eleanor Horne, Pat Kyllonen, Ernie Price, Larry Stricker, Rich Swartz,
Stan von Mayrhauser, Dylan Wiliam, and Ann Willard. However, all opinions contained herein, and
any errors of fact, are my own. I am also grateful for the production contributions of Bill Monaghan,
Bill Petzinger, and Sally Acquaviva, without whom this report would not have been possible.
iv
What Does It Mean to Be a Nonprofit
Educational Measurement Organization
in the 21st Century?
The philosopher George Santayana (1905) said,
“Those who cannot remember the past are condemned
to repeat it.” This quote is often called “Santayana’s
Warning,” because it is taken to mean that an understanding of history helps avoid having to relive
previous mistakes. But the quote can also be read
to suggest that, in order to make reasoned decisions
about the future, we need to be cognizant of where
we have come from. This claim is especially true
for a nonprofit organization because its continued
existence is usually rooted in its founding purposes.
This paper uses Educational Testing Service (ETS),
the largest of the nonprofit educational measurement
organizations, to illustrate the importance of remembering the past. The paper is divided into four sections.
First, the tax code governing the establishment and
operation of educational nonprofits is reviewed.
Second, the history around the founding of ETS is
described. Third, the implications of ETS’s past for
its future are discussed. Finally, the main points of
the paper are summarized.
What Is an Educational Nonprofit?
The term “nonprofit” refers to how an organization
is incorporated under state law. To be federally tax
exempt, an educational nonprofit must become a
501(c)(3) corporation.1,2 What is 501(c)(3)? It is a
very important section in the Internal Revenue Code.
The section is important because of what it does, and
does not, allow educational nonprofits to do, as well
as because of how the section came about.
Section 501(c)(3) exempts certain types of organizations from federal income tax.3 To qualify, an
organization must meet certain discrete “tests.”
The tests are the organizational test, operational test,
inurement test, lobbying restriction, electioneering
prohibition, public benefit test, and public policy test
(Harris, 2004). Each of these tests is briefly reviewed
in turn.
Under the Internal Revenue Code, to be exempt,
an organization must be set up exclusively for one or
more of the following purposes: charitable, religious,
educational, scientific, literary, testing for public
safety, fostering amateur national or international
sports competition, or the prevention of cruelty to
children or animals (Internal Revenue Service [IRS],
2003b). An entity meets this organizational test if its
articles of incorporation limit its function to one or
more exempt purposes (e.g., educational) and do not
expressly allow the organization to engage, other than
insubstantially, in activities that are not consistent
with those purposes. ETS’s exempt purpose is “educational,” and its organizing documents specify the
activities it can pursue in keeping with that purpose.
Paraphrasing the 2005 revision of the organization’s
Charter and Bylaws (Educational Testing Service
[ETS], 2005), those activities are as follows:
• conduct educational testing services;
• counsel test users on measurement;
• s erve as a clearinghouse about research
in testing;
• d etermine the need for, encourage, and carry on
research in major areas of assessment;
• p romote understanding of scientific educational
measurement and the maintenance of the highest
standards in testing;
• p rovide teachers, parents, and students (including adults) with products and services to improve
learning and decisions about opportunities;
• e nhance educational opportunities for minority
and educationally disadvantaged students; and
• e ngage in other advisory services and activities
in testing and measurement from time to time.
To meet the second — or operational — test, the
organization must be run exclusively for one or more
of the exempt purposes designated in its articles. The
test is met if the organization’s stated purpose and
activities conform. Although Section 501(c)(3)
indicates that the organization must be operated
1
For purposes of this paper, “nonprofit” and “501(c)(3) corporation” are used to mean the same thing, even though they are legally different.
2
The terms “nonprofit” and “not-for-profit” are not legally distinct, at least not in the Internal Revenue Code.
3
he IRS lists 27 types of organizations that are tax exempt under 501(c), only one type of which covers those institutions exempt under 501(c)(3)
T
(IRS, 2003a).
1
“exclusively” for exempt purposes, the term
“exclusively” has been interpreted by the IRS to
mean “primarily” or “substantially.” Thus, Section
501(c)(3) does allow exempt organizations to
engage in activities unrelated to their exempt
purposes (IRS, 2000). But those activities must not
become “substantial” and tax must generally be
paid on this unrelated business income.
Note that the operational test makes clear that
engaging in unrelated activities to support the exempt
purpose is, in itself, a non-exempt purpose, if it is
done any more than insubstantially (IRS, n.d.a.).4 To
prevent such unrelated activities from becoming so
substantial that they threaten tax-exempt status —
as well as to allow outside investment and limit
liability — an exempt organization may create
for-profit subsidiaries.5
The inurement test is often cited as the fundamental
difference between for-profit and nonprofit corporations. This test says that no part of the organization’s
net earnings may benefit any private individual. For
example, there may be no stockholders and no
distribution of net earnings, as in a dividend.
The lobbying restriction and electioneering prohibition mean, respectively, that no significant part of
an organization’s activities may consist of “carrying
on propaganda or otherwise attempting to influence
legislation ...” and that an exempt organization may
not participate or intervene in any political campaign
for or against any candidate for public office
(IRS, n.d.b.).6 Unlike the lobbying restriction,
the electioneering prohibition is absolute.
To meet the public benefit test, the organization
must operate for the advantage of public, rather than
2
private, interests.7 Private interests can be benefited,
but only incidentally. Further, the principal
beneficiaries of the organization’s activities must
be sufficiently numerous and well-defined so that
the community is, in some way, served.
Finally, there is the public policy test, which essentially says that an otherwise qualifying organization’s
“... purpose must not be so at odds with the common
community conscience as to undermine any public
benefit that might otherwise be conferred” (Bob
Jones University v. United States, 1983). The
quintessential example is Bob Jones University,
which lost its tax-exempt status as a result of racially
discriminatory practices that the IRS claimed, and
the Supreme Court affirmed, violated fundamental
public policy.8
Organizations set up for educational purposes
under 501(c)(3) have several additional requirements
(IRS, 2003b). First, the “positions” they take must be
educational. According to the IRS, “... advocacy of a
particular position ... may be educational if there is a
sufficiently full and fair exposition of pertinent facts
to permit an individual or the public to form an
independent opinion or conclusion” (IRS, 2003b,
p. 21). Also, the method used by an organization
to develop and present its views is a factor in
determining if the organization is “educational.”
What constitutes an “educational” method? The
IRS says that the method is not educational when:
• the presentation of viewpoints unsupported by
facts is a significant part of the organization’s
communications;
• the facts that purport to support the viewpoints
are distorted; or
4
here does not appear to be a statutory or regulatory definition of “substantial.” However, experts in nonprofit tax law often advise limiting gross
T
unrelated business income to no more than 10 to 20 percent of gross revenue (e.g., “FAQʼs - 501(c)(3) status,” n.d.).
5
The Chauncey Group International would be one example from ETSʼs history.
6
he dollar limits associated with the lobbying restriction are defined by a relatively complex formula. See “Restrictions on Nonprofit Activities for
T
501(c)(3) Organizations” (n.d.). Also see IRS (n.d.c.).
7
his test differs from the inurement test in that the inurement test applies only to insiders — persons having a private interest in the organization’s
T
activities — whereas the “private interests” cited in the public benefit test apply more generally.
8
he IRS revoked the tax-exempt status of Bob Jones University in 1975 even though the school had not violated any specific provision of 501(c)(3).
T
The IRS revoked the university’s tax-exempt status because the university, on the basis of religious belief, had at first refused admission to Black
students, and then narrowed that restriction to exclude from admission only Black students engaged in interracial marriage or who were known to
advocate such marriage or dating. It then admitted Black students generally but put in place strict consequences, including expulsion, for interracial
dating. The university contested the IRS decision and the case was eventually brought to the U.S. Supreme Court, which effectively upheld the IRS
decision by an 8–1 vote.
• the organization’s presentations express
conclusions more on the basis of emotion
than objective evaluation.
That, then, is what 501(c)(3) is about. But why did
Congress decide to grant tax exemptions to certain
organizations in the first place, thereby forgoing
huge amounts of future revenue? The statutory roots
of 501(c)(3) are commonly traced to the Tariff Act
of 1894, which imposed a corporate income tax and
exempted entities organized and conducted solely for
charitable, religious, or educational purposes from
having to pay it (Scrivner, 2001). The Congressional
intent behind the exemption was to give preferential
treatment because such organizations provided a
benefit to society. Congress reaffirmed this view in
the Revenue Act of 1938 when it said that tax exemption was based on the theory that the loss of revenue
is compensated for by relieving the government of a
function it would otherwise have to perform (presumably because the for-profit sector would not, or
should not be allowed to, perform it) and because of
the benefits to the general welfare that the function
would serve (Bob Jones University v. United States,
1983).9
The Revenue Act of 1950 added unrelated business
income tax rules, which were intended to eliminate
unfair competition by taxing the unrelated activities
of exempt organizations in the same way as competing for-profit corporations were taxed (Scrivner,
2001). The Internal Revenue Code of 1954 was a restructuring to the current numbering, which resulted
in the section known today as “501(c)(3).” Finally,
the 1959 Regulations for the 1950 Act and the 1954
Code defined “charity” to more closely approach the
English common-law definition (Scrivner, 2001) —
that is, not only the relief of poverty, but also the
advancement of education, religion, and other
purposes beneficial to the community. So legally,
ETS is, like many similar 501(c)(3) organizations,
a type of public charity.10
9
10
To summarize, in the words of the majority opinion
rendered by the U.S. Supreme Court in Bob Jones
University v. United States (1983), “In enacting ...
501(c)(3), Congress sought to provide tax benefits to
charitable organizations to encourage the development of private institutions that serve a useful public
purpose or supplement or take the place of public
institutions of the same kind.”
Thus, Section 501(c)(3) has its roots in the idea
that the government might not be able to provide
all the services the public needs, that the for-profit
sector might not fill the gap, and that those organizations which do voluntarily address such social needs
should be compensated through tax exemption.
How did ETS come to be a 501(c)(3)? The reasons
for that lie fundamentally in how ETS came about.
That story begins at the end of the 19th century, just
prior to the establishment of the College Entrance
Examination Board.
Where Did ETS Come From?
Prior to the founding of the College Entrance
Examination Board (CEEB), admission to colleges
and universities in the United States was a disorganized, if not chaotic, process (Fuess, 1950). The Ivy
League institutions each administered their own tests,
which varied widely in subjects assessed, quality,
and administration date. Wilson Farrand, principal
of Newark Academy, summarized the disarray in
entrance requirements as follows (cited in Fuess,
1950, p. 17): “Princeton requires Latin of candidates
for one course, but not for the others. Yale demands it
of all, Columbia of none. Princeton names five books
of Caesar and four orations of Cicero; Yale names
four books of Caesar and three books of Virgil ... Yale
calls for Botany, Columbia for Physics and Chemistry, Princeton for no science. Princeton and Columbia
demand both German and French, while Yale is satisfied with either. On the other hand, while Princeton
hy shouldn’t the for-profit sector supply some services? Because the need for profit may come into direct conflict with the intended public benefit
W
behind the service. Some services require a disinterested party. See, for example, the inurement test, the lobbying restriction, and the electioneering
prohibition, which are intended to distance the service provider from self-interest that could otherwise affect the provision of the service. Occupational
and professional licensing and certification, which is often handled by private nonprofit associations, would be an example.
ll 501(c)(3) organizations must be categorized under Section 509(a) as either private foundations or one of several types of public charity. See IRS
A
(2003b, pp. 37–40) for the particulars of this classification.
3
and Columbia demand only American History, Yale
calls also for that of England ...”
“Psychological” examinations, and it was led by Carl
Brigham of Princeton University.
Other colleges annually reviewed and certified
high schools so that they had a means of assuring the
quality of the curriculum and of the recommendations
coming from school principals (Fuess, 1950; Hubin,
1988). This certification system was a burden for
both the colleges and the high schools. College staff
had to physically visit each feeder school, and each
feeder school had to undergo multiple reviews annually, one for each receiving college. To help rationalize
admissions, the CEEB was established in 1900
through the Association of Colleges and Secondary
Schools of the Middle States and Maryland, with 12
collegiate charter members, as well as representation
from secondary schools (Fuess, 1950).11 The Board’s
initial purpose was to create a single battery of
centrally scored examinations and, in so doing, bring
order and higher quality to the college preparatory
curriculum. The first “College Boards,” as they were
called, were administered the following year, 1901.
As the Board was pursuing
its twin investigations, the
American Council on
Education (ACE) began its
own initiative to develop
“psychological examinations.” ACE was founded
in 1918, as an association
dominated by public
universities. (The muchsmaller CEEB, in contrast,
Edward L. Thorndike
was composed primarily
of private institutions.) In 1924, ACE commissioned
Louis. L. Thurstone, of the Carnegie Institute of
Technology, to create an admissions test based
on Army Alpha.13 ACE took this course of action
because many public institutions were already using
intelligence tests in admissions and ACE wanted to
standardize this use (Hubin, 1988).
The College Boards were a weeklong battery of
essay tests in various content domains, which solved
some problems but brought others to the fore (Hubin,
1988). By the early 1920s, the CEEB membership
was unhappy enough with the exams that it began
publicly voicing concerns about their subjectivity in
scoring, the variation in difficulty from one administration (or form) to the next, and the narrowing effect
the tests were having on the high school curriculum.12
As a consequence of this unhappiness, the Board
commissioned two streams of investigation, with
the idea being to supplement, not replace, the essay
tests (Hubin, 1988). The first stream of investigation
focused on multiple-choice achievement tests and
was led by Edward L. Thorndike and Ben D. Wood
of Teachers College, Columbia University.
Under the auspices of the
CEEB, Brigham had by
1926 developed his “psychological” examination,
but under a new name,
because Brigham, originally a eugenicist, no
longer believed that
intelligence tests measured
native ability. And in the
first Scholastic Aptitude
Ben Wood
Test (SAT®) manual
(Brigham, 1926, cited in Hubin, 1988, pp. 196–197),
he was quite clear:
The second stream centered on multiple-choice
intelligence tests derived from Yerkes’ 1918 Army
Alpha Test. The Board referred to this project as its
4
“The term ‘scholastic aptitude’ makes no stronger
claim ... than that there is a tendency for ... scores
in these tests to be associated positively with ...
subsequent academic attainment.”
11
he charter members were Barnard College, Bryn Mawr College, Columbia University, Cornell University, Johns Hopkins University, New York
T
University, Rutgers College, Swarthmore College, Union College, University of Pennsylvania, Vassar College, and the Womanʼs College of Baltimore
(Fuess, 1950).
12
These issues, incidentally, remain with us in one way or another to this day.
13
The Carnegie Institute would later merge with the Mellon Institute of Industrial Research to form Carnegie-Mellon University.
Further:
“This additional test ... should be regarded merely
as a supplementary record.”
To evaluate Brigham’s predictive validity claim,
the CEEB began to administer the test experimentally
that same year (Hubin, 1988).
The story next moves to Harvard. The year was
1933 and the country was in the midst of the Great
Depression. James Bryant Conant had just become
president of the university. As president, Conant
found that applicants were being drawn primarily
from a small number of northeastern preparatory
schools and that the “College Boards” were still
being used for admission (Hubin, 1988). Conant disliked the “College Boards” because he saw them as
nothing more than a measure of mastery of the prep
school curriculum that couldn’t be used to assess
students coming from public schools. For Conant,
Harvard admission was being based largely on ability
to pay. If a student could not afford to attend prep
school, that student was not going to do well on the
College Boards, and wasn’t coming to Harvard.
Conant decided to address
this problem by creating a
scholarship program to
increase economic and
regional diversity (Hubin,
1988). Note that Conant
did not want to increase
access to Harvard for
everyone, but just for those
with academic talent. How
would one measure acaJames Bryant Conant
demic talent in the public
schools? Obviously not with the College Boards,
which were keyed toward the prep school curriculum.
To find a solution, Conant turned to Henry Chauncey,
his assistant dean of admissions. Chauncey contacted
Carl Brigham, who was by that time Associate
Director for Research at the CEEB, on leave from
Princeton University. Chauncey arranged with the
CEEB for Brigham’s Scholastic Aptitude Test to be
used for the award of Harvard scholarships in the
very next year (1934). By 1937, the “Scholarship
Tests,” as they were now being called, were used by
14 colleges. The Scholarship Tests were composed of
two parts: the Scholastic Aptitude Test, as developed
by Brigham, and a battery of achievement tests
created by Ben Wood for ACE’s Cooperative Test
Service. This use of the Scholarship Tests was funded
by the Carnegie Foundation for the Advancement of
Teaching. In the following year, 1938, the Scholarship Tests were extended for use in university
admissions within a small group of institutions.
Collaborating in the Scholarship Test project
were the three organizations that would later form
ETS: the College Entrance Examination Board, the
American Council on Education, and the Carnegie
Foundation for the Advancement of Teaching. This
third organization, the Carnegie Foundation, was
established by Andrew Carnegie in 1905 as an independent policy and research center to “... encourage,
uphold, and dignify the profession of the teacher and
the cause of higher education” (Carnegie Foundation
for the Advancement of Teaching, 2005). Around
the time the Scholarship Tests were being put into
place, William Learned of the Foundation was also
encouraging development of the Graduate Record
Examinations® (GRE®) as an experimental test for
admission to graduate liberal arts programs (Hubin,
1988). The GRE consisted of an aptitude test, which
was originally the SAT, and a battery of achievement
tests created by Wood for ACE. The GRE was first
administered experimentally at Harvard, Yale, Prince­
ton, and Columbia in 1937, the very same year the
Scholarship Tests were extended beyond Harvard.
Just as the GRE and the Scholarship Tests were
being introduced, several key figures floated the idea
of a unified testing organization (ETS, 1992). In a
1937 speech at the annual meeting of the Educational
Records Bureau, Conant advocated the formation of a
“nationwide cooperative testing service,” because he
believed that standardized testing would be advanced
by a substantial research program concentrated in
one organization. Around the same time, Learned of
Carnegie approached Wood of ACE about forming
a “general examinations board,” because Learned
believed that testing could be made more efficient by
eliminating competition and duplication, and because
he thought that more resources would be available for
research and development if a unified organization
was formed (Hubin, 1988).
5
George Mullins, CEEB Secretary, described the
philosophical foundation that such an organization
should have (ETS, 1992, p. 9). He wrote:
“The organization should be built so that it can
attack the problems of educational measurement
scientifically ... It should have no doctrine to sell,
no propaganda concerning certain forms of tests to
be spread ... It must be an open-minded scientific
organization if it is destined to give real service
and therefore to endure.”
Carl Brigham
Clearly, scientific research
was a principal motivation
for the advocates of a unified
testing organization. And,
paradoxically, scientific
research was also a motivation for those opposed. The
strongest opponent was Carl
Brigham (ETS, 1992; Hubin,
1988; Lemann, 1999). What
were Brigham’s concerns?
First, Brigham did not believe that psychology or
measurement was scientifically advanced enough to
support the large-scale operational use of testing that
a national agency would bring. He wrote (cited in
Lemann, 1999, p. 34):
“Practice has always outrun theory ... this is a new
field and ... very little has been done which
is right.”
What was not right? A prime example comes from
an earlier letter (cited in Lemann, 1999, p. 33):
“The more I work in this field, the more I am
convinced that psychologists have sinned greatly
in sliding easily from the name of the test to the
functions or trait measured.”
This comment referred to the ease with which
the psychologists of the day had concluded that IQ
tests measured innate intelligence tied to ethnicity, a
view he himself had taken but since, quite publicly,
rejected (Hubin, 1988).
A second reason for Brigham’s opposition to a unified measurement organization was that he believed
6
consolidation would slow the growth of measurement science and kill innovation in tests. In a letter to
Conant (cited in ETS, 1992, p. 6), he wrote:
“One of my complaints against the proposed organization is that although the word research will
be mentioned many times in its charter, the very
creation of powerful machinery to do more widely
those things that are now being done badly will
stifle research, discourage new developments, and
establish existing methods, and even existing tests,
as the correct ones.”
What kind of innovation did Brigham wish to
see? Among other things, Brigham was interested in
connecting tests to teaching, learning, and cognition
(Donlon, 1979; Hubin, 1988). Ideas in his 1932 book,
A Study of Error, anticipated by a half-century what
we today call “formative assessment.” Brigham
believed that the mistakes students made in solving
test items could provide a basis for instruction.
But according to Lemann (1999), what concerned
Brigham most was that any organization that owned
the rights to a particular examination would inescapably become more interested in marketing that test
than in objectively researching its effectiveness and
working toward its improvement. For Brigham, a
strong research program, not just lip service, was essential because “The provision for extensive research
will prevent degeneration into a sales and propaganda
group ...” (Brigham, 1937, p. 756). Brigham’s opposition was so strident, and his opinion so respected, that
the idea for a consolidated organization was shelved
— that is, until Brigham died in January of 1943, at
the relatively young age of 52.
Coincidentally, by the time Brigham died, the need
for a unified testing agency had become more pronounced. In the 1940s, the CEEB was still primarily
a regional membership organization, though it had
grown from its original 12 institutions to more than
50 (Fuess, 1950).
While membership growth was certainly desirable,
the Board found it difficult to integrate the increased
operational testing activity it had to perform. The
SAT was now equated and machine scored, making
it both fairer and far more efficient to process than
the College Boards. In 1941, the old College Boards
were discontinued which, from William Learned’s
point of view, could have been no great loss, for
he had earlier described them as “a few arbitrarily
chosen questions with answers rated in terms of
personal opinion by individuals of varying degrees
of experience and competence” (cited in Hubin,
1988, p. 293).
operations, Law School Admission Test (LSAT)
operations, 139 permanent employees, and its
Princeton office. Carnegie gave the GRE, the PreEngineering Inventory, and 36 employees. ACE
donated the Cooperative Test Service, its National
Teacher Examinations, Thurstone’s Psychological
Examination, and 37 employees.14
The 1940s saw the CEEB’s operational testing
activities balloon. Initially, this growth came from
large military contracts during World War II, most of
which had nothing to do with educational testing and
which, consequently, made the Board’s membership
of admissions officers quite uneasy (Hubin, 1988).
This activity was followed by an upsurge in college
applicants after the war because of the GI Bill,
which paid essentially the full cost of college for
returning veterans.
In December 1947, ETS was granted a charter by
the New York State Board of Regents. James Conant
was designated chairman of the Board of Trustees.
Chauncey, now the director of the CEEB, was named
ETS’s first president, in recognition of the role he
played in bringing the three organizations together.
Meanwhile, the Carnegie Foundation had its own
concerns. In 1944, the GRE had gone operational
and, by 1946, the GRE unit had more staff than
the Foundation proper (Hubin, 1988). While the
Foundation had hoped for such success, it was in
the business of seeding development, not running
established testing programs so, like the Board, it
too was looking for a home for its “Big Test.”
Finally, ACE’s Cooperative Test Service (CTS) was
operating at a loss (Lemann, 1999). Moreover, ACE,
which saw the CEEB as a competitor, did not want
the CEEB to get the GRE. Thus, all three organizations had their own motivations for wanting a
centralized testing agency. Carnegie took the next
step by sponsoring a national commission, chaired
by Conant, to recommend how to proceed
(Bowles, 1967).
When the commission issued its recommendations,
there was considerable negotiation, much of it facilitated by Henry Chauncey, over whether the agency
should be a functionary of one of the three organizations or independent of them (ETS, 1992; Lemann,
1999). Finally, the three organizations agreed on an
independent arrangement in which they would turn
over their testing programs and a portion of their
assets to the new ETS (ETS, 1992; Fuess, 1950).
Among other things, the CEEB contributed SAT
14
Interestingly, neither the Board of Trustees, nor
Henry Chauncey, forgot Carl Brigham’s warnings
about the need for research as a mechanism for
driving innovation, helping maintain high standards,
and, ultimately, improving education. With respect to
improving education, after its first meeting, the ETS
Board of Trustees issued the following statement
(cited in ETS, 1992, p. 22):
“In view of the great need for research in all areas
and the long-range importance of this work to the
future development of sound educational programs,
it is the hope of those who have brought the ETS
into being that [through its research] it make
fundamental contributions to the progress of
education in the United States.”
Henry Chauncey also made the connection
between research and improvements to education.
In his inaugural message to staff (cited in ETS, 1992,
pp. 21–22), he wrote:
“It is our ardent hope and confident expectation
that the new organization will make important
contributions to American education through
developing and making available tests of the
highest standards, by sponsoring distinguished
research both on existing tests and on unexplored
test areas, and by providing effective advisory
services ... to schools and colleges.”
For Chauncey, these weren’t just words. During his
tenure as ETS president, traditions for the conduct
of research were established to forestall precisely the
Each organization also contributed capital in the form of cash or securities (Fuess, 1950).
7
problems that Brigham feared. Chauncey put into
place academic freedom to encourage independent
thinking and constructive criticism of ETS’s own
tests, thereby fostering opportunities for the instruments’ improvement. He put into place a policy
requiring full publication of research results, unless
the outcomes of a project were identified in advance
as proprietary and confidential. He established
this policy for two reasons. The first reason was to
prevent the suppression of results unfavorable to
ETS or its testing programs because, if suppression
occurred, it would signal the organization’s descent
into producing “propaganda” to serve its own ends.15
A second reason for the publication policy was that
Chauncey saw advancing the fields of education and
measurement as a corporate obligation, and publication as a mechanism for achieving it.
Third, Chauncey put into
place a funding source
that was purposely not
controlled by the testing
programs. He did this to
encourage long-term,
forward-thinking
projects unconstrained
by short-term business
needs. Last, he actively
encouraged a wideranging agenda to push
Henry Chauncey
innovation along many
measurement and education fronts simultaneously.
Collectively, Chauncey’s traditions had one other
important effect. They created an environment that
allowed him to attract and retain the brightest young
research staff — Frederic Lord, Samuel Messick,
Ledyard Tucker, Harold Gulliksen and Norman
Frederiksen — and, as a result, build ETS’s
reputation as a leader in scientific educational
measurement.
To summarize, Conant, Chauncey, Learned, and the
others who spearheaded the creation of ETS saw the
new organization as providing a public service
8
that neither the government nor the private sector
offered. In this way, the public service principles
underlying the founding of ETS and those of the
Internal Revenue Code met, with ETS being
established as a nonprofit corporation.
What Does the Past Imply for the Future?
Turning back to Santayana, how might an organization like ETS use its past, in particular its public
service roots, to help shape its future? In other
words, how can such an organization succeed as
a nonprofit educational measurement organization
in the 21st century?
To answer these questions requires a close look at
the idea of mission. Section 501(c)(3) doesn’t use
that word. It uses the term “purpose” to describe the
reason for which an exempt organization is operated.
What’s an organizational “mission?” It is, essentially,
nothing more than a statement of purpose. Logically then, a nonprofit organization’s “mission” must
be aligned with its exempt purpose; otherwise, the
organization could not remain exempt. ETS’s mission
is “... to help advance quality and equity in education
by providing fair and valid assessments, research and
related services,” which is quite close in meaning
to the inaugural words of the trustees and of Henry
Chauncey, as well as to ETS’s chartered purpose.16
As noted earlier, the overwhelming majority of
activities an exempt organization conducts must be
consistent with its purposes under section 501(c)(3).
Such an organization may not, as a primary operating
strategy, use unrelated business to fund its mission
(IRS, n.d.a.). So, what additional strategies are available to fund the mission for such an organization?
One strategy often used by educational nonprofits
with large endowments is to fund activities through
investment proceeds. But this approach is not a longterm operating strategy for an organization with relatively limited reserves. Such reserves are needed as a
contingency against unforeseen emergencies and as
an occasional source of support for special projects.
15
I t is interesting to note the repeated appearance of the word “propaganda” in the story of ETS: Brigham feared it, George Mullins (CEEB) hoped a
unified organization would transcend it, the IRS enjoined educational nonprofits against it, and Henry Chauncey made sure to do something about it.
16
TSʼs chartered purpose reads as follows (ETS, 2005): “The purposes for which the corporation is formed are to engage in, undertake, and carry on
E
services, research and other activities in the field of educational testing and such other activities as may be appropriate to such purposes.”
A second mechanism is through for-profit subsidiaries, which protect the exempt status of the parent
from unrelated income and allow outside investment.
This was an approach that ETS used in the 1990s
with the establishment of Chauncey Group International (CGI), K–12 Works, and ETS Technologies.
CGI was sold in 2004 to Thomson Prometric. The
other two subsidiaries were incorporated into ETS
proper a short time before. More recently, ETS has
created several new subsidiaries, including one
for-profit, ETS Global BV.
The third strategy is to fund the mission through
the mission — in essence, through new product
and service development, because the products
that originally funded the mission are mature and
cannot necessarily be counted on as a source of
long-term support.
There is, of course, a catch to funding the mission
through mission-related new product and service
development. Obviously, these new products and
services need to be financially successful. That’s not
the catch.
The catch is that, to be truly mission oriented, this
new product and service development needs to be
different in some fundamental way from the products
and services that for-profits offer. Otherwise, what
justifies nonprofit status? How can it be a public
service if the for-profit sector is offering the very
same thing?
How, then, does an organization like ETS
differentiate itself? ETS has typically tried to
differentiate itself by providing higher levels of
quality — or of integrity — than the for-profits
offer. And when people’s life chances are at stake,
that is a public service.
But there may be something more that an organization like ETS can do. That “something more”
is to take on the big challenges the for-profits will
not because the challenges are too hard and they
don’t have the scientific capability to attack them, or
because the return on investment isn’t big enough or
soon enough, and their need to satisfy shareholders
won’t allow it.
What are the big challenges? Arguably, the best
examples can be found in the work of ETS’s progenitors. Carl Brigham was interested in going beyond
the SAT’s selection function to using assessment as
a guide for instruction. And what better time to
pursue that goal than now, under No Child Left
Behind (NCLB)? The relevant questions today are:
• Who needs help?
• W
ith what specific content standards do they
need help?
• H
ow should they be taught those specific
standards?
These questions are especially important to address
in the case of minority and economically disadvantaged students, for whom closing the achievement
gap has become a critical national issue.
For his part, Henry Chauncey had enormous
interest in measuring new constructs to guide
individuals in educational choices (Lemann, 1999).
He wanted to know:
• What are individuals good at?
• What are they interested in?
• H
ow can those aptitudes and interests be guided
for the good of the individual and society?
Finally, James Conant’s original interest was using
assessment to increase diversity. In today’s terms:
• H
ow can we better locate promising individuals
from all backgrounds for purposes of encouraging them to pursue higher education?
• H
ow can we help them further develop the skills
they will need to succeed (including such new
basics as English for second-language learners)?
The challenges that Brigham, Chauncey, and
Conant posed well over 50 years ago remain largely
unresolved today. If a nonprofit measurement organization like ETS can create financially successful
new products and services that meet those challenges,
it will be doing as it should: funding its mission by
doing its mission.
9
But such an organization does not have a chance of
realizing this goal without reorienting dramatically.
For one, it must become more responsive to market
needs. That means:
• s cientific research and development that solves
problems customers care about;
• b usiness units that influence the choice of
problems to be solved and that help shape those
solutions into marketable products;
• s ales and marketing functions that effectively
communicate what differentiates a nonprofit’s
products from competitor offerings, and
that bring back intelligence about customer
needs, product ideas, and potential
product improvements;
• technology delivery that can readily and costeffectively incorporate advances; and
10
• o perational processes that execute accurately
and efficiently.
However, in addition to being more responsive,
such an organization must be more responsible.
That means:
• p roducts and marketing claims the organization
can stand behind scientifically;
• r esearch that supports, but is not controlled by,
the business units; and
• r esearch and development investments that are
determined not only by financial return but also
mission return (i.e., the positive impact on education the product or service is expected to have).
Ultimately, being “responsible” means continuing
to champion integrity, quality, and scientific leadership in educational measurement and in education.
Summary
ETS is a nonprofit educational measurement organization with public service principles deeply rooted
in its progenitors and in Section 501(c)(3). Both its
progenitors and the framers of the tax code were interested in the same thing: providing a social benefit.
A corollary of this history is that the purpose of a
nonprofit measurement organization like ETS is not
about making money. Its purpose is not even about
making money to support its mission (because
that, in itself, would be a non-exempt purpose). Its
purpose is about doing its mission, making enough
money in the process to continue doing it in better
and bigger ways.
many of these big challenges are the same ones that
motivated the organization’s progenitors. Those
individuals knew these were hard problems, but
they had faith that those problems could be solved.
To solve them, ETS and organizations like it will
need to act more responsively and more responsibly
than ever before.
Make no mistake: If ETS succeeds in being
responsive without being responsible, or in being
responsible without being responsive, it will have
failed. ETS, and kindred organizations, must do
both. Doing both is what it means to be a nonprofit
educational measurement organization in the
21st century.
Third, it is reasonable to argue that, in the 21st
century, renewed meaning can be brought to ETS’s
mission by attacking big challenges in education
and measurement, such as using assessment to
guide instruction, measuring new constructs to help
individuals with educational choices, and using
assessment to increase diversity. Interestingly,
11
References
Bob Jones University v. United States, 461 U.S. 574
(1983).
Bowles, F. (1967). The refounding of the College
Board, 1948–1963. New York: College Entrance
Examination Board.
Brigham, C. C. (1937). The place of research in
a testing organization. School and Society, 46,
756–759.
Carnegie Foundation for the Advancement of Teaching. (2005). About the Carnegie Foundation.
Retrieved January 26, 2005, from http://www.
carnegiefoundation.org/aboutus/index.htm
Donlon, T. F. (1979). Brigham’s book. The College
Board Review, 113, 24–30.
Educational Testing Service. (1992). The origins
of Educational Testing Service. Princeton, NJ:
Author.
Educational Testing Service. (2005). Educational
Testing Service: Charter and bylaws (Revised
April 2005). Princeton, NJ: Author.
FAQ’s – 501(c)(3) status. (n.d.). Retrieved January
26, 2005, from the Thompson & Thompson Web
site: http://www.t-tlaw.com/lr-04.htm
Fuess, C. M. (1950). The College Board: Its first fifty
years. New York: Columbia University Press.
Harris, J. A. (2004). Requirements for federal income
tax exemption under code section 501(c)(3).
Retrieved January 22, 2005, from http://www.
zsrlaw.com/publications/articles/jah1999.htm
Hubin, D. R. (1988). The Scholastic Aptitude Test:
Its development and introduction, 1900–1947.
Retrieved January 22, 2005, from the University
of Oregon Web site: http://darkwing.uoregon.
edu/~hubin/
12
Internal Revenue Service. (n.d.a). Unrelated
business income tax – General rules. Retrieved
January 22, 2005, from http://www.irs.gov/
charities/article/0,,id=96104,00.html
Internal Revenue Service. (n.d.b.). Section 501(c)(3)
organizations. Retrieved January 26, 2005, from
http://www.irs.gov/publications/p557/ch03.
html#d0e3351
Internal Revenue Service. (n.d.c.). Political and
lobbying activities. Retrieved January 26, 2005,
from http://www.irs.gov/charities/charitable/
article/0,,id=120703,00.html
Internal Revenue Service. (2000). Tax on unrelated
business income of exempt organizations (IRS
publication 598). Retrieved January 22, 2005,
from http://www.irs.gov/pub/irs-pdf/p598.pdf
Internal Revenue Service. (2003a). IRS data books:
2001 – 2003. Retrieved January 22, 2005, from
http://www.irs.gov/pub/irs-soi/01db22eo.xls
Internal Revenue Service. (2003b). Tax exempt
status for your organization (IRS Publication No. 557). Retrieved January 22, 2005,
from http://www.irs.gov/pub/irspdf/p557.
pdf#search=’irs%20publication%20557
Lemann, N. (1999). The big test: The secret history
of the American meritocracy. New York: Farrar,
Strauss and Giroux.
Restrictions on nonprofit activities for 501(c)(3)
organizations. (n.d.). Retrieved January 26, 2005,
from http://www.alliance1.org/Public_Policy/
IRS_rules.htm
Santayana, G. (1905). The life of reason, reason in
common sense. New York: Scribner’s.
Scrivner, G. N. (2001). A brief history of tax policy
changes affecting charitable organizations. In
J. S. Ott (Ed.), The nature of the nonprofit sector
(pp. 126–142). Boulder, CO: Westview.
Copyright © 2011 by Educational Testing Service. All rights reserved. ETS, the ETS logo, LISTENING. LEARNING. LEADING., GRADUATE RECORD EXAMINATIONS and
GRE are registered trademarks of Educational Testing Service (ETS). College Board and SAT are registered trademarks of the College Board. 16452